Ivan from a Tech Room
Region: DE
Thursday 27 August 2026 16:33:07 GMT
52423
1817
81
121
Music
Download
Comments
software_training :
for 100% of people I would suggest buy a $20 subscription and wait for 2028
2026-08-30 16:04:17
39
knudvaneeden :
Interesting to know why people were suddenly buying this Apple mini. Because of their very special hardware design thus out of the box.
2026-09-05 01:42:35
0
Mylife :
Yes, 13 token/sec for basically every dense model. It will fit, but it will barely run
2026-08-28 22:25:49
3
Tym@r :
...There is several mini pcs with unified memory
2026-08-28 11:36:34
1
Chairman 125 :
DGX Spark is better
2026-08-30 20:12:33
4
Frank :
You can also use Ryzen AI Max CPU based Systems to get the same advantage.
2026-08-27 16:44:56
4
Stoapiperl :
The bandwidth is so slow. Much smarter to get a 700€ 7900xtx which runs qwen 3.8-27b at 56t/s
2026-08-27 19:57:05
10
cumshoodaboiu23 :
You have to hope the models even run on Mac chips
2026-08-27 19:57:25
4
0_0hector :
there are another options with unified memory more affordable and better (nvidia spark and ryzen ai max as example)
2026-08-29 07:30:00
1
Itiswhatitizz :
What can I run smoothly on my MacBook Pro M5 pro 24 Gb ?
2026-08-28 21:07:36
0
qwerty :
Stupid expensive for slow tokens on big models
2026-08-28 21:50:25
5
Henry :
Just wait for Xiomi AI machine
2026-08-30 14:30:20
6
Антон от Creativiso :
are 16 gb enough?
2026-08-28 05:26:03
0
Snowflake :
Buy an old Gen 3 board for xeon 22cores with 6 ram slots, install the 5070ti and run Gemma 4 26b under 40c without any lag and instant responses!
2026-08-27 18:05:26
0
mr.mjau1 :
I have a 6 month old server(PC) that have good hardware, but a really bad graphic card. What should I buy to be able to experiment with the new Qwen-modell? Looking forward to next episode
2026-08-27 16:54:18
1
Layetano :
From 4000$ and up…
2026-08-28 08:00:19
1
RainMan :
well, they are expensive thay should run faster.. i prefer AMD AI MAX
2026-08-30 12:58:06
2
Daboon :
I rather spend more and buy pc
2026-08-31 10:52:17
1
AppleArabTech :
The problem is very expensive to upgrade, while in PC you can add another GPU or more ram or more storage in way less
2026-08-31 15:32:22
1
JL :
But you cannot run nvidia cuda .
2026-08-28 23:17:03
2
Filippo :
Mac mini m1 i think is not enough
2026-08-27 19:43:51
0
Filipe Tarouca :
🍀💪🙏
2026-08-29 00:08:02
1
WonderCandaule :
xiaomi is here now
2026-08-28 05:27:09
2
Petersilie am Zahn :
Even if someone would give me an up2date Apple device for free, I would deny.
2026-08-28 17:18:04
1
/home/StachuJones :
But Local Models works better on Linux
2026-08-28 01:03:56
1
To see more videos from user @easybits.tech.room, please go to the Tikwm
homepage.
